Abstract
The world has come to a sudden halt due to the incessant spread of a viral pneumonia dubbed
COVID-19 caused by the beta-coronavirus, SARS-CoV-2. The pandemic spread of the virus
has already claimed lakhs of valuable lives and has infected millions of people across the globe.
The situation is further worsened by the fact that there is no approved therapeutics currently
available for the treatment of the disease. The only way to handle the crisis is the rapid
development of a therapeutic strategy to combat the virus. Computational biology offers
resources to rapidly identify novel drug leads and to repurpose existing drugs at the expense of
minimal resources and time. The main protease of SARS-CoV-2 is key to the replication and
propogation of the virus in the host cells. Inhibiting the protease blocks replication and hence
is an attractive therapeutic target in the virus. The crystal structures of the protein in complex
with inhibitors are available in public databases. Here we describe the screening of the
DrugBank database, a public repository for small molecule therapeutics, to identify approved
or experimental phase drugs that can be repurposed against the main protease of SARS-CoV2. The initial screening was performed on more than 13,000 drug entries in the target database
using an energy optimised pharmacophore hypothesis AARRR. A sub-set of the molecules
selected based on the fitness score was further screened using molecular docking by
sequentially filtering the molecules through the high throughput virtual screening, extra
precision and standard precision docking modalities. The most promising hits were subjected
to binding free energy estimation using the MMGBSA method. Approved drugs viz, Cobicistat,
Larotrectinib and Simeprevir were identified as potential candidates for repurposing. Drugs in
the discovery phase identified as inhibitors include the known cysteine protease inhibitors,
Calpain inhibitor IV and an experimental cathepsin F inhibitor.